Large Content And Behavior Models To Understand, Simulate, And Optimize Content And Behavior

TL;DR

This paper introduces Large Content and Behavior Models (LCBMs) that incorporate receiver behavior tokens into training to optimize content for desired behaviors and predict receiver responses, extending beyond traditional content understanding.

Contribution

The paper presents the novel concept of LCBMs that integrate receiver behavior data into language models, enabling behavior prediction and content optimization.

Findings

Models perform similarly to LLMs on content understanding tasks.

Models generalize well for behavior simulation and domain adaptation.

The release of the Content Behavior Corpus (CBC) supports further research.

Abstract

Shannon and Weaver's seminal information theory divides communication into three levels: technical, semantic, and effectiveness. While the technical level deals with the accurate reconstruction of transmitted symbols, the semantic and effectiveness levels deal with the inferred meaning and its effect on the receiver. Large Language Models (LLMs), with their wide generalizability, make some progress towards the second level. However, LLMs and other communication models are not conventionally designed for predicting and optimizing communication for desired receiver behaviors and intents. As a result, the effectiveness level remains largely untouched by modern communication systems. In this paper, we introduce the receivers' "behavior tokens," such as shares, likes, clicks, purchases, and retweets, in the LLM's training corpora to optimize content for the receivers and predict their behaviors. Other than showing similar performance to LLMs on content understanding tasks, our trained models show generalization capabilities on the behavior dimension for behavior simulation, content simulation, behavior understanding, and behavior domain adaptation. We show results on all these capabilities using a wide range of tasks on three corpora. We call these models Large Content and Behavior Models (LCBMs). Further, to spur more research on LCBMs, we release our new Content Behavior Corpus (CBC), a repository containing communicator, message, and corresponding receiver behavior (https://behavior-in-the-wild.github.io/LCBM).